Simple and Complex Models

Attempts to model the urban development process have traditionally led to increasingly complex models which attempt to consider the myriad interactions between a region's land uses, demographic and economic changes, transportation system, land prices and rents, and public policies.(1) Contemporary examples include UrbanSim (Waddell, 2002), TRANUS (de la Barra, 2001), and LEAM (Deal and Pallathucheril, 2008), which are reviewed with several other examples by Timmermans (2006). More recently, a number of GIS-based planning support systems (PSS) have abandoned the attempt to duplicate a complex reality for much simpler modeling approaches. State-change models such as SLEUTH (Clarke, 2008), CUF II, and CURBA (Landis, 2001) project future growth patterns by calibrating the model to observed land-use changes and assuming that these changes will continue into the future. Rule-based models such as What if? (Klosterman, 1999; 2008) assume that the user's assumptions about the relative suitability of different locations, the overall trend of future development, and the effectiveness of alternative public policies will prove to be correct. This commentary considers the advantages and disadvantages of simple and complex models for urban and regional planning and highlights the critical role that assumptions play in attempts to forecast the future. It concludes by suggesting that the choice between these modeling approaches reflects more fundamental assumptions about the limits of science, the role of the public, and the nature of planning.

[1]  Keith C. Clarke,et al.  A Decade of Cellular Urban Modeling with SLEUTH: Unresolved Issues and Problems , 2008 .

[2]  Michèle Hibon,et al.  Accuracy of Forecasting: An Empirical Investigation , 1979 .

[3]  Steve Bankes,et al.  Exploratory Modeling for Policy Analysis , 1993, Oper. Res..

[4]  Martin Wachs.  Ethical Dilemmas in Forecasting for Public Policy , 1982 .

[5]  Richard E. Klosterman,et al.  The What If? Collaborative Planning Support System , 1999 .

[6]  S K Smith,et al.  Evaluating the forecast accuracy and bias of alternative population projections for states. , 1992, International journal of forecasting.

[7]  S K Smith,et al.  Further thoughts on simplicity and complexity in population projection models. , 1997, International journal of forecasting.

[8]  G. Box,et al.  Empirical Model-Building and Response Surfaces. , 1990 .

[9]  Martin Wachs,et al.  Forecasting versus Envisioning: A New Window on the Future , 2001 .

[10]  Richard E. Klosterman,et al.  The politics of computer-aided planning , 1987 .

[11]  O. Wolkenhauer.  Why model? , 2013, Front. Physiol..

[12]  Andrew M. Isserman,et al.  Population forecasting and local economic planning: The limits on community control over uncertainty , 1984 .

[13]  William H. Starbuck,et al.  Innocents in the Forest: Forecasting and Research Methods , 1990 .

[14]  P. Waddell.  UrbanSim: Modeling Urban Development for Land Use, Transportation, and Environmental Planning , 2002 .

[15]  R. L. Henshel,et al.  Forecasting: An Appraisal for Policy-Makers and Planners. , 1978 .

[16]  H. Baum,et al.  Forgetting to Plan , 1999 .

[17]  William Alonso,et al.  PREDICTING BEST WITH IMPERFECT DATA , 1968 .

[18]  S. Smith,et al.  Tests of forecast accuracy and bias for county population projections. , 1987, Journal of the American Statistical Association.

[19]  John F. Long,et al.  Complexity, accuracy, and utility of official population projections , 1995 .

[20]  J. Scott Armstrong,et al.  Beyond Accuracy: Comparison of Criteria Used to Select Forecasting Methods , 1995 .

[21]  William Ascher,et al.  The forecasting potential of complex models , 1981 .